CN114428334A

CN114428334A - Seismic data static correction processing method and device, computer equipment and storage medium

Info

Publication number: CN114428334A
Application number: CN202011042565.8A
Authority: CN
Inventors: 亢永敢; 魏嘉; 陈金焕; 朱海伟; 庞锐
Original assignee: China Petroleum and Chemical Corp; Sinopec Geophysical Research Institute
Current assignee: China Petroleum and Chemical Corp; Sinopec Geophysical Research Institute
Priority date: 2020-09-28
Filing date: 2020-09-28
Publication date: 2022-05-03
Also published as: WO2022062508A1

Abstract

The invention provides a seismic data static correction processing method, a device, computer equipment and a storage medium, wherein the method comprises the steps of obtaining target seismic data; inputting the target seismic data into a pre-trained initial picking neural network model for training to obtain first-arrival picking data containing first-arrival time data; inputting the first arrival picking data containing the first arrival time data into a static correction processing neural network for training to obtain static corrected first arrival picking data and second first arrival time data; calculating to obtain a static correction value according to the second first-arrival time data and the first-arrival time data after static correction processing; and correcting the seismic data according to the static correction value. And the automatic first arrival picking and the direct static correction calculation are realized. The method does not need to manually pick up first arrival data, avoids a complex near-surface modeling process, and realizes a high-efficiency and accurate static correction processing function.

Description

Seismic data static correction processing method and device, computer equipment and storage medium

Technical Field

The invention relates to the technical field of seismic data processing, in particular to a seismic data static correction processing method, a seismic data static correction processing device, computer equipment and a storage medium.

Background

The static correction is an important link of seismic data processing, and whether the static correction processing is accurate or not is directly related to the effect of a series of subsequent processing. The static correction processing method widely applied at present is to perform near-surface velocity tomography modeling processing by using first-arrival data to obtain near-surface velocity model data, and calculate travel time difference caused by undulating surface by using a velocity model to perform static correction. Accurate first arrival data and near-surface velocity model data are required. The first arrival picking process is time-consuming and labor-consuming, the near-surface speed modeling process is complex, and the acquisition of an accurate near-surface speed model is a difficult process. In the face of seismic data of complex surface exploration areas such as mountainous regions, accurate first-arrival data and near-surface velocity models need to be acquired very difficultly, and therefore the static correction effect is influenced. Static correction processing methods for complex earth surface seismic data are multiple, and mainly focus on first arrival pickup and high-precision near-earth surface velocity modeling, for example, an automatic first arrival pickup method, a chromatography near-earth surface velocity modeling method and the like are widely researched and applied, a certain processing effect is achieved, but the static correction problem of complex earth surfaces such as mountainous regions and the like cannot be completely solved. How to obtain the accurate static correction processing effect is a difficult problem for seismic exploration in complex mountainous regions.

Disclosure of Invention

In view of the above, it is necessary to provide a seismic data statics correction processing method, apparatus, computer device and storage medium for solving the above technical problems.

A seismic data static correction processing method comprises the following steps:

acquiring target seismic data;

inputting the target seismic data into a pre-trained initial picking neural network model for training to obtain first-arrival picking data containing first-arrival time data;

inputting the first arrival picking data containing the first arrival time data into a static correction processing neural network for training to obtain static corrected first arrival picking data and second first arrival time data;

calculating to obtain a static correction value according to the second first-arrival time data and the first-arrival time data after static correction processing;

and correcting the seismic data according to the static correction value.

In one embodiment, the step of inputting the target seismic data into a pre-trained initial picking neural network model for training to obtain first arrival picking data including first arrival time comprises:

acquiring sample data in a preset format, wherein the sample data in the preset format is each channel of shot gather data, the format of the seismic channel data comprises a file header, channel header data of each channel and a channel data body, and the channel data body records an amplitude value on each sampling point;

and inputting the sample data in the preset format to an initial picking neural network for training to obtain the initial picking neural network model.

In one embodiment, the step of obtaining sample data in a preset format further includes:

the method comprises the steps of constructing an initial picking neural network comprising a first input layer, a first intermediate layer and a first output layer, wherein the first input layer is used for inputting a piece of seismic data, the first output layer outputs a piece of data which is consistent with the number of sampling points of the input seismic data after calculation through the first intermediate layer, and each sampling point value in the output data is a first sampling point value or a second sampling point value.

In one embodiment, the step of inputting the target seismic data into a pre-trained initial picking neural network model for training to obtain first arrival picking data including first arrival time data comprises:

inputting the target seismic data into a pre-trained initial picking neural network model for training, wherein the initial picking neural network model outputs the initial picking data containing sample values, and the sample values comprise a first sample value and a second sample value;

extracting the first arrival picking data with the sample point value as the first sample point value through the first arrival picking neural network model, and acquiring the first arrival time data of the first arrival picking data corresponding to the first sample point value.

In one embodiment, the step of inputting the first arrival pickup data including the first arrival time data into a static correction processing neural network for training, and obtaining the static correction processed first arrival pickup data and second first arrival time data includes:

inputting the first arrival picking data containing the first arrival time data into a static correction processing neural network for training;

and resetting the statically corrected elevation of the target seismic data needing the static correction processing through the static correction processing application network, and outputting the first arrival time of the target elevation.

In one embodiment, the step of inputting the first arrival pickup data including the first arrival time data into a static correction processing neural network for training, and obtaining the static corrected first arrival pickup data and second first arrival time data further includes:

and constructing a static correction processing neural network comprising a second input layer, a second middle layer and a second output layer.

In one embodiment, the step of calculating a static correction amount according to the second first arrival time data and the first arrival time data after the static correction processing includes:

and calculating the difference between the second first-arrival time data and the first-arrival time data after static correction processing to obtain the static correction value.

A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor when executing the computer program implements the steps of:

acquiring target seismic data;

and correcting the seismic data according to the static correction value.

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

acquiring target seismic data;

and correcting the seismic data according to the static correction value.

The seismic data static correction processing method, the seismic data static correction processing device, the computer equipment and the storage medium realize first arrival automatic pickup and direct static correction calculation. The method does not need to manually pick up first arrival data, avoids a complex near-surface modeling process, and realizes a high-efficiency and accurate static correction processing function.

Drawings

FIG. 1 is a schematic diagram of an application scenario of a seismic data static correction processing method in one embodiment;

FIG. 2 is a schematic flow chart diagram illustrating a seismic data statics processing method according to one embodiment;

FIG. 3 is a block diagram of a seismic data statics correction processing apparatus in one embodiment;

FIG. 4 is a diagram illustrating an internal structure of a computer device according to an embodiment;

FIG. 5 is a schematic diagram illustrating an implementation of a seismic data statics processing method in one embodiment;

FIG. 6 is a diagram of first break pickup effects in one embodiment;

FIG. 7A is a diagram illustrating first arrival picking before static correction processing in one embodiment;

FIG. 7B is a diagram illustrating first arrival picking after static correction processing in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

Example one

The seismic data static correction processing method provided by the application can be applied to the application environment shown in FIG. 1. Wherein the computer 102 communicates with the server 104 over a network. The terminal 102 may be, but not limited to, various personal computers, servers, laptops, smartphones, tablets and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers. A user sends target seismic data to a server 104 through a terminal 102, and the server 104 acquires the target seismic data; inputting the target seismic data into a pre-trained initial picking neural network model for training to obtain first-arrival picking data containing first-arrival time data; inputting the first arrival picking data containing the first arrival time data into a static correction processing neural network for training to obtain static corrected first arrival picking data and second first arrival time data; calculating to obtain a static correction value according to the second first-arrival time data and the first-arrival time data after static correction processing; and correcting the seismic data according to the static correction value.

Example two

In this embodiment, as shown in fig. 2, a seismic data static correction processing method is provided, which includes:

step 210, target seismic data is acquired.

Step 220, inputting the target seismic data into a pre-trained initial picking neural network model for training, and obtaining first arrival picking data containing first arrival time data.

Specifically, a data volume of the sample seismic data is extracted from the target seismic data, and file header description information of the sample seismic data and header information of each channel are removed. And verifying the acquisition of the data body, namely generating the data body with the same size according to the seismic data body, setting each sampling point value in the data body to be 0, and then setting the value of the sampling point position corresponding to the first arrival time of each track to be 1 to represent the first arrival time position of the track of data.

Extracting one track of data once from the acquired seismic data volume, inputting the amplitude value of each sampling point of one track as input data into each node of the initial picking neural network model, and taking the track data of the meaning first arrival time corresponding to the input track as output inspection data. When the seismic data are trained, the seismic data are not input in sequence, but one data is input at a certain distance, and the data are circulated in sequence, so that the training of all input data is finally completed.

Specifically, a seismic channel data body is extracted from seismic data needing first arrival picking, and the seismic channel data body is input into a channel initial picking neural network model according to the channel sequence. And after the calculation of the initial pickup neural network model, taking the sampling position of the seismic channel corresponding to the node with the node value of 1 of the output layer as the first arrival time of the target channel. And sequentially inputting all seismic data needing first arrival picking into the channel nerve completion, and finally acquiring the first arrival time of all the data.

The method comprises the steps of inputting a piece of seismic data to an initial picking neural network model, outputting data with the number of sampling points consistent with that of the input seismic data by an output layer after calculation of the initial picking neural network model, wherein each sampling point value in the data is 1 or 0. 1 indicates that the sample point is a first arrival time point, and 0 indicates that the sample point is a non-first arrival time point. In this embodiment, the first-arrival time point corresponding to the sampling point is first-arrival time data, and the first-arrival time data is first-arrival time obtained by picking up the actual elevation.

Step 230, inputting the first arrival picked data including the first arrival time data into a static correction processing neural network for training, and obtaining the static corrected first arrival picked data and a second first arrival time number.

The goal of the static correction processing neural network is to establish the relationship between the surface elevation and the first arrival time. The neural network structure is divided into an input layer, an intermediate layer and an output layer. And inputting the coordinates and the elevation of the shot point position of the horizon seismic channel data and the coordinates and the elevation of the demodulator probe position. Therefore, the input layer is provided with six nodes, and shot point coordinates and elevations (Sx, Sy and Sz) and geophone point coordinates and elevations (Rx, Ry and Rz) corresponding to each channel of data are distributed. The middle layer is provided with two layers, and the number of nodes of each layer is 50. And inputting a node of the layer and outputting the first arrival time. The neural network is a fully connected network.

In this embodiment, the shot point coordinates and elevations (Sx, Sy, Sz) and the geophone point coordinates and elevations (Rx, Ry, Rz) of each track of the seismic data are extracted, the extracted six parameters are input to an input layer of the neural network, and the first arrival time of the input channel is used as verification data of the output layer. And sequentially carrying out neural network training on each track of the target seismic data.

And 240, calculating to obtain a static correction value according to the second first-arrival time data and the first-arrival time data after the static correction processing.

In one embodiment, the step of calculating a static correction amount according to the second first arrival time data and the first arrival time data after the static correction processing includes: and calculating the difference between the second first-arrival time data and the first-arrival time data after static correction processing to obtain the static correction value.

And establishing a relation between each path of elevation data and the first arrival time by the selected neural network, resetting the statically corrected elevation of the target seismic data needing the static correction processing, inputting the statically corrected elevation into the neural network, and calculating and outputting a value as the first arrival time of the target elevation through the neural network. And subtracting the first arrival time of the target elevation from the first arrival time of the actual elevation pickup to obtain a difference value, namely the static correction value of the target elevation.

And step 250, correcting the seismic data according to the static correction value.

And acquiring static correction of the target elevation for each channel of data according to the operation, and finishing static correction processing.

In the above embodiment, the first arrival automatic pickup and the direct static correction calculation are realized. The method does not need to manually pick up first arrival data, avoids a complex near-surface modeling process, and realizes a high-efficiency and accurate static correction processing function.

acquiring sample data in a preset format, wherein the sample data in the preset format is each channel of shot gather data, the format of the seismic channel data comprises a file header, channel header data of each channel and a channel data body, and the channel data body records an amplitude value on each sampling point; and inputting the sample data in the preset format to an initial picking neural network for training to obtain the initial picking neural network model.

Specifically, after the design of the deep neural network structure is completed, the neural network model needs to be trained by using sample data to obtain an initial pick-up neural network model. The sample data adopts accurate first arrival time data obtained by manual picking. The input sample is each channel of shot gather data, and the format of the seismic channel data comprises a file header, channel header data of each channel and a channel data body. The track data volume records the amplitude value of each sampling point, and the neural network input data only needs the track data volume, so that the input seismic data needs to be reconstructed, the file header description data and the track header data of each track are stripped, the track data volume of each track is reserved, and sample data of a pure data volume is formed. And the output sample data is constructed according to the input sample data and the first arrival time. And generating an empty data body with the same size according to the size of the input sample data, wherein the value of each sampling point in the data body is 0. And setting the value of a corresponding sampling point of each channel of data in the output sample data as 1 according to the first arrival time to obtain the output sample data. The training process of the neural network is as follows: and inputting the generated input sample data into an output layer of the neural network according to the channel sequence and inputting one channel of data each time, wherein each sampling point of the input channel data corresponds to a node of an input layer. The output data is output sample data of the corresponding track. The sampling point of each output sample corresponds to a node of one output layer.

In this embodiment, according to the characteristics of the seismic data, the initial pickup neural network is provided with an input layer, an intermediate layer, and an output layer. And the sampling point number of the node data of the input layer is consistent with that of the seismic data, and the node data of the input layer is used for inputting the seismic data. The middle layer is two layers, the number of nodes of each layer is consistent with that of the input layer, and the number of nodes of the output layer is consistent with that of the input layer. The network is a fully connected network. And inputting a path of seismic data by the input layer, outputting data consistent with the number of sampling points of the input seismic data by the output layer after calculation of the intermediate layer, wherein each sampling point value in the data is 1 or 0. 1 indicates that the sample point is a first arrival time point, and 0 indicates that the sample point is a non-first arrival time point.

Specifically, first, a first arrival pick-up neural network structure is set. A neural network structure is designed according to an input layer, two intermediate layers and an output layer, the node data of the input layer is consistent with the sampling point number of seismic data needing first arrival picking, and the node data of the two intermediate layers and the node data of the output layer are consistent with the node data of the input layer. Subsequently, training sample data is generated. And selecting part of seismic data needing first arrival picking as training sample data, and firstly, manually picking up first arrivals of the training sample seismic data to obtain accurate first arrival time. And extracting a data body of the sample seismic data according to the neural network structure, and removing file header description information of the sample seismic data and header information of each channel. And verifying the acquisition of the data body, namely generating the data body with the same size according to the seismic data body, setting each sampling point value in the data body to be 0, and then setting the value of the sampling point position corresponding to the first arrival time of each track to be 1 to represent the first arrival time position of the track of data. Then, training parameters are determined. The training parameters of the neural network are key factors for determining the training effect, and the training of the neural network is controlled by two parameters, namely the cycle number and the error amount, in consideration of the calculation amount and the precision of the training. The training error determines the precision of the training and prevents overfitting. The cycle number controls the calculation amount of training, and prevents the training from falling into multiple cycles and being unable to finish normally.

inputting the target seismic data into a pre-trained initial picking neural network model for training, wherein the initial picking neural network model outputs the initial picking data containing sample values, and the sample values comprise a first sample value and a second sample value; extracting the first arrival picking data with the sample point value as the first sample point value through the first arrival picking neural network model, and acquiring the first arrival time data of the first arrival picking data corresponding to the first sample point value.

In this embodiment, a trained neural network model is used, target seismic data that needs to be subjected to first arrival picking processing is generated into input data according to the requirements of input samples, a piece of data output after calculation by the neural network in a trained neural network of an input channel is input, a sampling point with a sampling value of 1 in an output channel is extracted, and the position time of the sampling point with the value of 1 is obtained as a first arrival picking result of the channel.

inputting the first arrival picking data containing the first arrival time data into a static correction processing neural network for training; and resetting the statically corrected elevation of the target seismic data needing the static correction processing through the static correction processing application network, and outputting the first arrival time of the target elevation.

Specifically, the shot coordinates and elevations (Sx, Sy, Sz) of each pass, and the geophone coordinates and elevations (Rx, Ry, Rz) are extracted for the seismic data for which first arrivals have been picked up. And respectively inputting the obtained six data of each track into six nodes of an input layer of the neural network, and outputting verification data as the initial time of input. And extracting training samples from all the picked first arrival seismic channels according to the processing procedure to read the neural network for training.

In one embodiment, the step of inputting the first arrival pickup data including the first arrival time data into a static correction processing neural network for training, and obtaining the static correction processed first arrival pickup data and second first arrival time data further includes:

Specifically, the goal of the static correction processing neural network is to establish the relationship of surface elevation to first arrival time. The neural network structure is divided into an input layer, an intermediate layer and an output layer. And inputting the coordinates and the elevation of the shot point position of the horizon seismic channel data and the coordinates and the elevation of the demodulator probe position. Therefore, the input layer is provided with six nodes, and shot point coordinates and elevations (Sx, Sy and Sz) and geophone point coordinates and elevations (Rx, Ry and Rz) corresponding to each channel of data are distributed. The middle layer is provided with two layers, and the number of nodes of each layer is 50. And inputting a node of the layer and outputting the first arrival time. The neural network is a fully connected network.

In the above embodiment, the deep neural network is used to automatically pick up the first arrival of the seismic data, and after the first arrival data is acquired, the deep neural network is used to process the first arrival data to acquire the relationship between the first arrival time and the surface elevation. And calculating the first arrival time of the corresponding elevation by setting different target surface elevations according to the relationship between the first arrival time and the surface elevation, thereby realizing direct static correction processing.

It should be understood that, although the steps in the flowchart of fig. 2 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in fig. 2 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

EXAMPLE III

The method utilizes the deep neural network to automatically pick up the first arrival of the seismic data, and after the first arrival data is obtained, the deep neural network is utilized to process the first arrival data to obtain the relation between the first arrival time and the earth surface elevation. And calculating the first arrival time of the corresponding elevation by setting different target surface elevations according to the relationship between the first arrival time and the surface elevation, thereby realizing direct static correction processing. The direct static correction processing procedure is as follows: setting a uniform ground surface elevation, calculating first arrival time of the uniform elevation by using a deep neural network, subtracting the first arrival time of the uniform elevation from the actually picked first arrival time to obtain a time correction value of the target elevation, and directly performing static correction processing by using the time correction value.

In this embodiment, the seismic data static correction processing process is as follows:

(1) first arrival picking neural network structure

According to the characteristics of the seismic data, the initial pickup neural network is provided with an input layer, an intermediate layer and an output layer. And the sampling point number of the node data of the input layer is consistent with that of the seismic data, and the node data of the input layer is used for inputting the seismic data. The middle layer is two layers, the number of nodes of each layer is consistent with that of the input layer, and the number of nodes of the output layer is consistent with that of the input layer. The network is a fully connected network. And inputting a path of seismic data by the input layer, outputting data consistent with the number of sampling points of the input seismic data by the output layer after calculation of the intermediate layer, wherein each sampling point value in the data is 1 or 0. 1 indicates that the sample point is a first arrival time point, and 0 indicates that the sample point is a non-first arrival time point.

(2) First arrival pickup sample data acquisition

After the deep neural network structure design is completed, the neural network model needs to be trained by using sample data. The sample data adopts accurate first arrival time data obtained by manual picking. The input sample is each channel of shot gather data, and the format of the seismic channel data comprises a file header, channel header data of each channel and a channel data body. The track data volume records the amplitude value of each sampling point, and the neural network input data only needs the track data volume, so that the input seismic data needs to be reconstructed, the file header description data and the track header data of each track are stripped, the track data volume of each track is reserved, and sample data of a pure data volume is formed. And the output sample data is constructed according to the input sample data and the first arrival time. And generating an empty data body with the same size according to the size of the input sample data, wherein the value of each sampling point in the data body is 0. And setting the value of a corresponding sampling point of each channel of data in the output sample data as 1 according to the first arrival time to obtain the output sample data. The training process of the neural network is as follows: and inputting the generated input sample data into an output layer of the neural network according to the channel sequence and inputting one channel of data each time, wherein each sampling point of the input channel data corresponds to a node of an input layer. The output data is output sample data of the corresponding track. The sampling point of each output sample corresponds to a node of one output layer.

(3) First arrival pickup data acquisition

Generating input data according to the target seismic data needing first arrival picking processing by using the trained neural network model according to the requirements of input samples, inputting a piece of data output after calculation of the neural network in the trained neural network of a channel, extracting sampling points with sampling values of 1 in an output channel, and obtaining the position time of the sampling points with the sampling values of 1 as the first arrival picking result of the channel.

(4) Static correction processing neural network architecture setup

(5) Training sample generation

Shot coordinates and elevations (Sx, Sy, Sz) for each pass, and geophone coordinates and elevations (Rx, Ry, Rz) are extracted for seismic data that have picked up first arrivals. And respectively inputting the obtained six data of each track into six nodes of an input layer of the neural network, and outputting verification data as the initial time of input. And extracting training samples from all the picked first arrival seismic channels according to the processing procedure to read the neural network for training.

(6) Static correction process

And establishing a relation between each path of elevation data and the first arrival time by the selected neural network, resetting the statically corrected elevation of the target seismic data needing the static correction processing, inputting the statically corrected elevation into the neural network, and calculating and outputting a value as the first arrival time of the target elevation through the neural network. And subtracting the first arrival time of the target elevation from the first arrival time of the actual elevation pickup to obtain a difference value, namely the static correction value of the target elevation. And acquiring static correction of the target elevation for each channel of data according to the operation, and finishing static correction processing.

The invention provides a seismic data static correction method based on a deep neural network, which realizes direct static correction processing of seismic data, avoids processing processes such as first arrival pickup and near-surface velocity modeling, meets the requirements of complex surface seismic data static correction processing, improves the seismic data processing effect, reduces exploration cost and improves economic benefits.

Example four

Referring to fig. 5, in the first step, the first arrival picking neural network structure is set. A neural network structure is designed according to an input layer, two intermediate layers and an output layer, the node data of the input layer is consistent with the sampling point number of seismic data needing first arrival picking, and the node data of the two intermediate layers and the node data of the output layer are consistent with the node data of the input layer.

And secondly, generating training sample data. And selecting part of seismic data needing first arrival picking as training sample data, and firstly, manually picking up first arrivals of the training sample seismic data to obtain accurate first arrival time. And extracting a data body of the sample seismic data according to the neural network structure, and removing file header description information of the sample seismic data and header information of each channel. And verifying the acquisition of the data body, namely generating the data body with the same size according to the seismic data body, setting each sampling point value in the data body to be 0, and then setting the value of the sampling point position corresponding to the first arrival time of each track to be 1 to represent the first arrival time position of the track of data.

And thirdly, determining training parameters. The training parameters of the neural network are key factors for determining the training effect, and the training of the neural network is controlled by two parameters, namely the cycle number and the error amount, in consideration of the calculation amount and the precision of the training. The training error determines the precision of the training and prevents overfitting. The cycle number controls the calculation amount of training, and prevents the training from falling into multiple cycles and being unable to finish normally.

And fourthly, training a neural network. And according to the structure and parameters of the neural network, acquiring the seismic data volume in the second step, extracting one track of data once, inputting the amplitude value of each sampling point of one track as input data into each node of the input layer of the neural network, and taking the track data of the meaning first arrival time corresponding to the input track as output inspection data. When the seismic data are trained, the seismic data are not input in sequence, but one data is input at a certain distance, and the data are circulated in sequence, so that the training of all input data is finally completed.

And fifthly, picking up the first arrival. And (4) extracting seismic channel data volume from the seismic data needing first arrival pickup according to the requirements of the second step by using the neural network trained in the fourth step, and inputting the seismic channel data volume into the channel neural network according to the channel sequence. And after the neural calculation is finished, taking the sampling position of the seismic channel corresponding to the node with the node value of 1 of the output layer as the first arrival time of the target channel. And sequentially inputting all seismic data needing first arrival picking into the channel nerve completion, and finally acquiring the first arrival time of all the data. The picked up first arrival image is shown in fig. 6.

And sixthly, statically correcting and processing the neural network structure setting. The static correction neural network structure is divided into an input layer, an intermediate layer and an output layer. The input layer is provided with six nodes, and shot point coordinates and elevations (Sx, Sy and Sz) and wave detection point coordinates and elevations (Rx, Ry and Rz) corresponding to each channel of data are distributed. The middle layer is provided with two layers, and the number of nodes of each layer is 50. And inputting a node of the layer and outputting the first arrival time. The neural network is a fully connected network.

And seventhly, training a static correction neural network. And extracting the shot point coordinates and elevations (Sx, Sy and Sz) and the geophone point coordinates and elevations (Rx, Ry and Rz) of each path of the seismic data, inputting the extracted six parameters into an input layer of the neural network, and taking the first arrival time of the input channel as verification data of an output layer. And sequentially carrying out neural network training on each track of the target seismic data.

And step eight, performing static correction processing. And saving the trained neural network parameters. Determining the elevation of the earth surface according to the static correction requirement, replacing the elevation of the shot point and the elevation of the wave-building point in the seismic data channel needing the static correction according to the new elevation of the earth surface, acquiring the coordinates (Sx, Sy and Sz) of the shot point and the coordinates (Rx, Ry and Rz) of the wave-detecting point after the static correction, inputting the six acquired parameters of the new elevation into a channel neural network, and calculating and outputting the first arrival time data of the corresponding channel through the neural network. And subtracting the first arrival time obtained by calculating the target elevation from the first arrival time obtained by picking up the actual elevation to obtain a difference value, namely the static correction value. And sequentially calculating all the track data to obtain a final static correction value. And correcting the target seismic data according to the obtained static correction value to realize static correction processing. Referring to fig. 7A and 7B, the first-arrival picking is performed after the first-arrival picking static correction processing before the static correction processing.

EXAMPLE five

In this embodiment, as shown in fig. 3, a seismic data statics correction processing apparatus is provided, including:

a target seismic data acquisition module 310, configured to acquire target seismic data;

a first neural network training module 320, configured to input the target seismic data into a pre-trained initial picking neural network model for training 330, to obtain first-arrival picking data including first-arrival time data;

a second neural network training module 340, configured to input the first arrival pickup data including the first arrival time data into a static correction processing neural network for training, so as to obtain statically corrected first arrival pickup data and second first arrival time data;

a static correction value calculation module 350, configured to calculate a static correction value according to the second first arrival time data and the first arrival time data after static correction processing;

and the static correction module 360 is used for correcting the seismic data according to the static correction value.

In one embodiment, the seismic data statics processing apparatus further comprises:

the seismic trace data acquisition module is used for acquiring sample data in a preset format, wherein the sample data in the preset format is each trace of shot gather data, the format of the seismic trace data comprises a file header, trace header data of each trace and a trace data body, and the trace data body records an amplitude value on each sampling point;

and the initial picking neural network model training obtaining module is used for inputting the sample data in the preset format into an initial picking neural network for training to obtain the initial picking neural network model.

the initial picking neural network building module is used for building an initial picking neural network comprising a first input layer, a first intermediate layer and a first output layer, wherein the first input layer is used for inputting a piece of seismic data, after calculation through the first intermediate layer, the first output layer outputs a piece of data which is consistent with the sampling point number of the input seismic data, and each sampling point value in the output data is a first sampling point value or a second sampling point value.

In one embodiment, the second neural network training module includes:

a first arrival picking data input unit for inputting the first arrival picking data containing the first arrival time data into a static correction processing neural network for training;

and the target elevation output unit is used for resetting the statically corrected elevation of the target seismic data needing the static correction processing through the static correction processing application network and outputting the first arrival time of the target elevation.

and the static correction processing neural network construction module is used for constructing a static correction processing neural network comprising a second input layer, a second middle layer and a second output layer.

In an embodiment, the static correction amount calculation module is further configured to calculate a difference between the second first-arrival time data and the first-arrival time data after the static correction processing, so as to obtain the static correction amount.

For specific limitations of the seismic data static correction processing device, reference may be made to the above limitations on the seismic data static correction processing method, which are not described herein again. The units in the seismic data static correction processing device can be wholly or partially realized by software, hardware and a combination thereof. The units can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the units.

EXAMPLE six

In this embodiment, a computer device is provided. The internal structure thereof may be as shown in fig. 4. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program, and is deployed with a database for storing a first arrival pick-up neural network model and a static correction processing neural network model. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used to communicate with other computer devices. The computer program is executed by a processor to implement a seismic data statics processing method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the architecture shown in fig. 4 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, there is provided a computer device comprising a memory storing a computer program and a processor implementing the following steps when the processor executes the computer program:

step 210, target seismic data is acquired.

In one embodiment, the processor, when executing the computer program, further performs the steps of:

EXAMPLE seven

In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored, the computer program realizing the following steps when executed by a processor:

step 210, target seismic data is acquired.

In one embodiment, the computer program when executed by the processor further performs the steps of:

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A seismic data static correction processing method is characterized by comprising the following steps:

acquiring target seismic data;

and correcting the seismic data according to the static correction value.

2. The method of claim 1, wherein the step of inputting the target seismic data into a pre-trained initial pickup neural network model for training to obtain first arrival pickup data comprising first arrival times is preceded by:

3. The method of claim 2, wherein the step of obtaining sample data in a preset format further comprises:

4. The method of claim 3, wherein the step of inputting the target seismic data into a pre-trained initial pickup neural network model for training comprises the step of obtaining first arrival pickup data comprising first arrival time data comprising:

5. The method of claim 1, wherein the step of inputting the first arrival pickup data including the first arrival time data into a neural network for training, and obtaining the first arrival pickup data after the static correction and the second arrival time data comprises:

6. The method of claim 1, wherein the step of inputting the first arrival pickup data including the first arrival time data into a neural network for training, and obtaining the statically corrected first arrival pickup data and second arrival time data further comprises:

7. The method according to any one of claims 1 to 6, wherein the step of calculating a static correction amount based on the second first arrival time data and the first arrival time data after the static correction process comprises:

8. A seismic data static correction processing apparatus, comprising:

the target seismic data acquisition module is used for acquiring target seismic data;

the first neural network training module is used for inputting the target seismic data into a pre-trained initial picking neural network model for training to obtain first-arrival picking data containing first-arrival time data;

the second neural network training module is used for inputting the first arrival pickup data containing the first arrival time data into a static correction processing neural network for training to obtain the first arrival pickup data and second first arrival time data after static correction processing;

the static correction value calculation module is used for calculating to obtain a static correction value according to the second first-arrival time data and the first-arrival time data after the static correction processing;

and the static correction module is used for correcting the seismic data according to the static correction value.

9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.